Received-SPF: pass (google.com: domain of linux-kernel+bounces-164936-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Tue, 30 Apr 2024 22:10:43 -0700
From: Christoph Hellwig <hch@infradead.org>
To: John Hubbard <jhubbard@nvidia.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Jason Gunthorpe <jgg@nvidia.com>,
	LKML <linux-kernel@vger.kernel.org>, linux-rdma@vger.kernel.org,
	linux-mm@kvack.org, Mike Marciniszyn <mike.marciniszyn@intel.com>,
	Leon Romanovsky <leon@kernel.org>,
	Artemy Kovalyov <artemyko@nvidia.com>,
	Michael Guralnik <michaelgur@nvidia.com>,
	Alistair Popple <apopple@nvidia.com>,
	Pak Markthub <pmarkthub@nvidia.com>
Subject: Re: [RFC] RDMA/umem: pin_user_pages*() can temporarily fail due to
 migration glitches
Message-ID: <ZjHO04Rb75TIlmkA@infradead.org>
References: <20240501003117.257735-1-jhubbard@nvidia.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240501003117.257735-1-jhubbard@nvidia.com>

> +		pinned = -ENOMEM;
> +		int attempts = 0;
> +		/*
> +		 * pin_user_pages_fast() can return -EAGAIN, due to falling back
> +		 * to gup-slow and then failing to migrate pages out of
> +		 * ZONE_MOVABLE due to a transient elevated page refcount.
> +		 *
> +		 * One retry is enough to avoid this problem, so far, but let's
> +		 * use a slightly higher retry count just in case even larger
> +		 * systems have a longer-lasting transient refcount problem.
> +		 *
> +		 */
> +		static const int MAX_ATTEMPTS = 3;
> +
> +		while (pinned == -EAGAIN && attempts < MAX_ATTEMPTS) {
> +			pinned = pin_user_pages_fast(cur_base,
> +						     min_t(unsigned long,
> +							npages, PAGE_SIZE /
> +							sizeof(struct page *)),
> +						     gup_flags, page_list);
>  			ret = pinned;
> -			goto umem_release;
> +			attempts++;
> +
> +			if (pinned == -EAGAIN)
> +				continue;
>  		}
> +		if (pinned < 0)
> +			goto umem_release;

This doesn't make sense.  IFF a blind retry is all that is needed it
should be done in the core functionality.  I fear it's not that easy,
though.